I am attracted to a view of a priori evidence that relates (modal) imagination and intuition in something like the following way: modal imagination provides a "situational platform" within which intuitive judgments are made. I want to use the notion of imagination here very loosely, so that I can imagine impossible situations (e.g., that there is a square round ball on my desk). Intuitions are then used to determine whether or not the imagined situation is possible and/or whether or not some other proposition is true in that situation. Since I allow that you can imagine impossible situations (in my weak sense of imagine; perhaps "entertain" or "consider" would be better), imagining is not evidential. What is evidential is the associated intuition(s). The view is compatible with the Chalmers/Yablo view that (positive) conceivability is a guide to possibility, but it views conceivability as nonbasic evidence. For conceivability will be cashed out in terms of imagination-cum-intuition and intuition will count as the basic form of evidence.
But taking intuition as evidentially basic doesn't imply that there is no philosophical interest in conceivability. Specifically, in order for intuition to be basic evidence, it must at least be generally reliable. I will assume that this is so. But there will be specific cognitive circumstances where intuition is plausibly unreliable. (Cf., perception is generally reliable, but there will be specific cognitive circumstances where it is unreliable.) Now Bealer has articulated a number of those circumstances (e.g., lack of mental focus, theory-directed intution, etc.). But one question the Chalmers taxonomy of conceivability opens up is the possibility that there will be nonobvious cases where intuitions are unreliable and which need to be made clear by explicit philosophical investigation. Perhaps this is what Dave meant in an earlier post by saying that rational intuition is a blunt instrument.
I should add here that, in some respects, Bealer doesn't deny intuition simpliciter is a blunt instrument (since, as noted above, there are many sorts of situations where he would think intuition might be unreliable). The question being raised here is whether or not in the core cases in which Bealer does think that intuition is reliable, further fine-graining might not provide an even more solid epistemic foundation for moderate rationalism.