[...] More than once I've spent over two weeks on a single move in a critical position that I wanted to be absolutely sure I got right.
[...]
xarxziux's point about sometimes wishing to think longer for one move is a valid one especially for the most serious correspondence chess players.
In my experience, here on chess.com on the one hand side, the thing that can be used in such cases is vacation time.
On the other hand side, as opposed to computer engine based correspondence chess, games here are of course more on the lighter side focusing more on fun than on scientifically working out the very last details of every position, so typically doing without extra thinking time is a real option for most cases even for top players in games with 3 days per move.
I don't think that ratings need be considered a factor in this. Online ratings, even in blitz, are a fairly rough measure of player's strength, and especially so in online/correspondence chess. It is quite obvious that the way different people approach this form of chess here on chess.com varies greatly: from almost casual, short-calculation play which centered on the idea of being able to play one move whenever you get to find a brief moment off your other occupations, to actually treating it like a true correspondence-chess challenge, with accuracy of play even exceeding that of serious OTB games. All of these varieties of players participate in various online ("correspondence-like") chess events and everyone has to come to terms with the innate inadequacies of these encounters.
I think that the crux of the discussion is: which is more important -- the sporting aspect of the leagues (which definitely wants tighter schedules) or the big participation, the sort of bring-them-all-in approach? To me, both hold merit, but I probably lean towards the latter. Nonetheless, I think that Rehubelent's idea for a potential reform is a very commendable initiative.