In some piece of fiction (I think it was Orion’s Arm, but the closest I can find is http://www.orionsarm.com/eg-topic/45b3daabb2329 and the reference to “the Herimann-Glauer-Yudkowski relation of inclusive retrospective obviousness”) I saw the idea that one could order qualitatively-smarter things on the basis of what you’re calling “clicks”. Specifically, that if humans are level 1, then the next level above that is the level where if you handed the being the data on which our science is built, all the results would click immediately/be obvious.
I’ve seen it asserted that humans are essentially “Turing complete” with respect to intelligence; anything that can be understood by any intelligence can be understood by a non-broken human, given enough time and attention. I’m on the fence about that, frankly; there’s a lot of stuff that I have real trouble understanding, despite being decently bright by most standards. But if there’s such a thing as real quantitative (rather than qualitative) differences in intelligence, it seems to me that “clicks” are at the core of what such a thing would look like from the outside (not what it would be internally; no idea about that, of course).