Approximately correct. To be precise, by “Darwinian” mechanisms Behe usually has in mind primarily random mutations filtered by natural selection (a combination sticklers might say is more “neo-Darwinian” than Darwinian, since Darwin didn’t talk much about “randomness”), and he usually explicitly focuses on those, but he also indicates that other purported mechanisms of evolution, to the extent that they are unguided and unplanned, would fall under the same criticism that he levels against “Darwinian” mechanisms. But roughly speaking, I’d agree with what you are saying.
Yes, but always the context of those “waiting times” (waiting for the right random mutations) is what he calls “Darwinian” evolution; the book would have been more accurately titled “The Edge of Darwinian Evolution.”
Denton had the same problem with his “Evolution: A Theory in Crisis” book; the title makes it sound as if Denton thought that common descent was in doubt, but, as he clarifies in his third book, that is not what he meant, and the first book would have been more accurately titled “Darwinism: A Theory in Crisis” (see page 111). Of course, publishers aiming at popular or semi-popular audiences have a financial interest in giving their books short, catchy, and not-too-academic-sounding titles. “The Edge of Darwinian Evolution” sounds very academic, and so does “Darwinism: A Theory in Crisis”, so publishers aiming at a broad audience are going to sacrifice accuracy for marketability. This doubtless explains why Discovery, despite Denton’s explanation in his third book that he had never meant to challenge common descent, chose to stick with just plain “Evolution” in the title of the third book. So “Evolution” is “Still a Theory in Crisis”, even though what Denton meant in the third book was that the classical neo-Darwinian formulation of evolution is still a theory in crisis (to the point where in its pure form it is hardly held by anyone anymore, being seriously modified, according to Denton, by conceptions such as evo-devo).
The same point applies to your comment about adaptive evolution, which again in Behe assumes the (neo-)Darwinian context. That is, according to Behe, if the only thing going on were Darwinian mechanisms, we would expect degeneration of existing structures rather than the building of new ones. But if there is design involved anywhere in the evolutionary process (whether front-loaded at the beginning or inserted later), we would be in a whole new ball game.
The same applies to your summary statement, “what Behe thinks evolution can do isn’t all that much,” and for the same reason.
Some might protest that if “design” is introduced into the mix of causes, we are no longer talking about “evolution” as Darwin (with his naturalistic insistence) intended it (he did eventually adopt the term “evolution”), and I concede the historical point regarding Darwin; but the term “evolution” has been in the popular language for so long now that for many lay people it means primarily “descent with modification” from earlier, simpler forms", and within that framework, one can speak of Behe and Denton as affirming evolution. Indeed, if we denied the term “evolution” to any scheme of descent with modification in which there are genuine teleological factors, we would no longer be able to say that Lamarck or Bergson affirmed “evolution,” which seems to me to be an unnecessary, and even silly, restriction of language. I see Darwin, Lamarck, Bergson, Mayr, Gould, Dawkins, Coyne, Wagner, Behe, Denton, Shapiro, etc. as all affirming “evolution” while having quite a range of views regarding its causes.