Verne is a fairly strong example. He was reasonably popular in his time, so he didn't become famous only because of correlation-in-hindsight. We might ask whether he's an outlier or just the tail of the bell curve, but I strongly suspect that what actually happened was that later engineers were consciously influenced by his fictional designs, much like with William Gibson and the modern Internet.

This suggests that fiction-future correlation is largely determined by technical plausibility, which in turn suggests that we may be able to predict in advance which science-fiction predictions are most likely to come true. However, it occurs to me that we do a lot of this last already on this website (cryonics, uploading, nanotech, AI, computronium, ...) so I'm not sure if this quite counts as an "advance" strength of the theory.

(Other predicted technologies I don't remember seeing so much around here: Dyson spheres, space elevators, ... hm. Not that many, actually. Augmented reality and bionic implants are likely only transitional, but will probably have at least a few years of massive popularity at some point.)