As far as Themistocles's journey after his exile is concerned why do modern historians conclude that Plutarch meant Thasos instead of Naxos?

These 'modern scholars' actually explicitly explain their reasoning, if they firmly choose one port, which is not always the same consensus many might want to find here: the extant (ancient) manuscripts are not that many, but still divided about the issue and not uniformly pointing towards a simple solution. For reasons of internal plausibility the scholars opt a more consistent solution in reconstructing the events described.

So we seem to not see that 'modern scholars have uncovered the true meaning of Plutarch's words', but instead a critical re-evaluation of the sources we have, weighing Plutarch against Thucydides and interpreting the results?

To ask the opposite question: 'why preferring Naxos?':

The unequivocal decision to only mention Naxos, like in Wikipedia, in the Loeb version of Plutarch from 1914, or the Penelope version for Thucydides may rather be an attempt in modern times to harmonise the readings in favour of the interpretations?
The very same thing that scholars assume may have occurred when ancient historians tried to write their accounts or later copiest transmitted them.

Rhodes explains:

Some scholars have tried to exploit Themistocles’ flight across the Aegean. According to Thucydides (I. 137. ii) he set out in a merchant ship from Pydna in Macedon; he was travelling incognito, but when they came close to Naxos while the Athenians were besieging it he revealed himself to the captain and asked to be kept safe; and he eventually reached Ephesus. Plutarch (Them. 25. ii–26. i) claims to be following Thucydides, but takes Themistocles from Pydna past Thasos (probably: the manuscripts are divided between Thasos and Naxos) to Cyme. If we knew which siege Themistocles had to avoid, that would help us to date his crossing of the Aegean – but I suspect that the two versions of the story are rival embroideries on the fact that, when crossing the Aegean, he had to take care not to fall into the Athenians’ hands.

— P. J. Rhodes: "A History of the Classical Greek World 478–323 BC", Wiley-Blackwell: Malden, Oxford: 2010, p40.

Rahe explains:

Themistocles’ flight, its timing, his arrival in Anatolia, news of Xerxes’ death: Charon of Lampsacus FGrH 262 F11, Thuc. 1.136.1–137.3, Nep. Them. 8.2–9.1, Diod. 11.55.4–56.4, Plut. Them. 24.1–26.1, P olyaen. Strat. 1.30.7, Aristodemus FGrH 104 F8.2–5 with Robert Flacelière, “Sur quelques points obscurs de la vie de Thémistocles,” REA 55:1–2 (January–June 1953): 5–28; Philip Deane, Thucydides’ Dates, 465–431 B.C. (Don Mills, Ontario: Longman Canada, 1972), 9–13; Forrest, “Pausanias and Themistokles Again,” 115–20; Marcus P. Milton, “The Date of Thucydides’ Synchronism of the Siege of Naxos with Themistokles’ Flight,” Historia 28:3 (3rd Quarter, 1979): 257–75; Ron K. Unz, “The Chronology of the Pentekontaetia,” CQ n.s. 36:1 (1986): 68–85 (at 69–73); Victor Parker, “The Chronology of the Pentecontaetia from 465 to 456,” Athenaeum 81 (1993): 129–47 (a t 130–33); Marr, Commentary, 142–55; W. Kendrick Pritchett, “Thucydides’ Pentekontaetia: 5. Naxos, Thasos, and Themistokles’ Flight,” in Pritchett, Thucydides’ Pentekontaetia and Other Essays (Leiden: Brill, 1995), 81–94; Briant, CA, 563; Keaveney, The Life and Journey of Athenian Statesman Themistocles, 23–26, 114–16; a nd Green, “Commentary,” 119–20, n. 213. Death of Hiero ca. 467: Diod. 11.38.7, 66.4.

If I follow Flacelière, Forrest, Pritchett, Marr, and Briant, rather than Deane, Milton, Unz, Parker, Keaveney, and, I presume, Green in preferring the testimony of the best manuscript of Plutarch to that of the surviving manuscripts of Thucydides and in supposing that Themistocles slipped past the Athenians besieging Thasos, not Naxos, it is because I believe that what we can surmise concerning the chronology of this period requires it.

The sequence of events in Thucydides is clear. The revolt of Naxos preceded the battle of Eurymedon, which came before the revolt of Thasos. The first of these three events began at some point after 476.

Deane, Milton, Unz, and Parker to the contrary notwithstanding, it is not likely that these three events were bunched together in and after 466; and, Keaveney to the contrary notwithstanding, the revolt of Naxos is not likely to have continued for more than two or three years, if that. Eurymedon is likely to have taken place in 469, the year before Cimon and his fellow generals were asked to judge the tragic competition; and the beginning of the revolt at Thasos can be dated to 465, the year Xerxes is known to have died.

— Paul Anthony Rahe: "Sparta's First Attic War: The Grand Strategy of Classical Sparta, 478–446 B.C", Yale University Press: New Haven, London, 2019, p259.

The events apparently even cannot be reconstructed on watertight timeline, making inferences less sharp than would be desirable.

There remains, or rather we have exchanged for the delay, the problem of Themistokles' escape from the Athenian fleet besieging Naxos (Thuc. i I37.2). The date proposed by the editors of The Athenian Tribute Lists and now generally favoured for the revolt of Naxos is 470; but Themistokles, if he fled from Argos some years after 470, obviously could not have run into the siege of Naxos if it was going on in that year. This problem admits of no easy solution; but in the various attempts to find a solution two points have been raised. First, how reliable is this date (or for that matter the proposed alternative of 467) for the revolt of Naxos? Secondly, was it the siege of Naxos Themistokles narrowly escaped, or the siege of Thasos, as one-and by general consent the best-manuscript of Plutarch's Themistoklesreads at 25.2?

Flaceliere has argued persuasively that Θασον is the genuine reading in Plutarch, that there were two traditions about Themistokles' route, Naxos-Ephesos and Thasos-Kyme, and that Plutarch chose the latter as geographically a more natural route from Pydna and chronologically less difficult, since the revolt of Thasos occurred not long before the accession of Artaxerxes.

He believes that the reaδing Ναξον in the other manuscripts of Plutarch was a correction of an original Θασον tο bring Plutarch's account into conformity with that of Thucydides and Charon of Lampsakos. The counter-suggestion has been made: that the text of Thucydides originally read Thasos rather than Naxos. But it is not easy to see why a scribe copying the text of Thucydides should have altered the much easier reading Thasos to Naxos. Moreover the Naxos–Ephesos route was found in the account of the flight written by Thucydides' contemporary, Charon of Lampsakos (Plut. Them.27. I), and was followed later by Nepos (Them.8.6-9. I) and the Epistles.

— Mary E. White: "Some Agiad Dates: Pausanias and His Sons", The Journal of Hellenic Studies, Vol. 84 (1964), pp. 140-152 jstor