Specifically on the point you mention (incommensurability), I don't believe he recanted but clarified or refined his definition of the term:
Most readers of my texxt has supposed that when I spoke of theories as incommensurable, I meant that they could not be compared. But 'incommensurability' is a term borrowed from mathematics, and it there has no such implication. The hypotenuse of an isosceles right triangle is incommensurable with its side, but the two can be compared to any required degree of precision. What is lacking is not comparability but a unit of length in terms of which both can be measured directly and exactly. In applying the term 'incommensurability' to theories, I had intended only to insist that there was no common language within which both could fully expressed and which could therefore be used in a point-by-point comparison between them."
- Kuhn, "Hermeneutics and Social Science," Cultural Hermeneutics 2 (1975):312
Concerning Paradigms, Kuhn regretted the use of the term, but that was because of the ambiguity. More caustic critics claimed that the structure of scientific revolutions contained dozens of meanings for the word 'paradigm' (I seem to recall someone saying it got close to the triple digits). For his own part, Kuhn admitted to carelessly using the term in a few places, which opened up the work for extreme interpretations.
The subject is well-treated in Richard J. Bernstein's "Science, Rationality, Incommensurability" in Beyond Objectivism and Relativism (1983). Read the excellent work if you get a chance and this question genuinely interests you.