That's exactly my point Uri. My standard rating has no relevance to my playing strength. All it shows is that I am relatively good at solving Chess Tempo problems (and that I invest significantly more time in each problem than the average user). I have no illusions that I'm the 7th best tactician (my ranking as of this time) on CT as such.
What I use CT for, is practicing calculation training. Each problem I try to calculate all possible lines to the fullest extent, delve into the specific position and understand it fully. I have other programs I prefer for practicing pattern recognitions, easy problems, etc (stuff that some other users do here).
What works for me here is the added incentive of the ratings, which I find just give me that little extra to try and look that little bit harder, compared to when I'm just sitting with a book.
I am, and have always been a lousy blitzer. I just cannot think fast enough, and I doubt I ever will. I don't have any interest in it (although I enjoy the odd blitz game as much as the next guy). I suck at it and neither CTS or CT blitz has ever done any good for my abilities (ok, I never used CT in blitz mode but I doubt it would). that's not to say that other people won't benefit - that's just the way my brain seems to work.
Fortunately for me, in a real long game you
do have time to sit and analyse positions for 5-10 minutes, or even longer. Of course the key is to recognise when to use the extra time

, but you do not necessarily need to
see the tactics fast, seeing the
potential for tactics in a given situation is often enough (for me).
Incidently, the only rating I've ever had corresponds to roughly 2000 fide. Wouldn't put too much into that, as it was only based on 30 or so games.
On the other hand, if the site gets too perfect and becomes exactly like a real game, you might as well just go play a real game instead. As revenant points out, it's sometimes good to train certain things, rather than always be in a real-game situation.
CT is never going to become like a real game, even if it included all kinds of different tactics. How many games do you know that you have to find one single good tactical move in each position from move 1 (or in 90-95% of the positions if there was no-solution problems)?.
As in all aspects of life, splitting up your training is not only sometimes good, it's a must. I've done my thousands of fork-problems, mate in Xs, etc all stuff that has helped me improve to where I am today. However these things done individually can only take you so far. There's still a big jump to a real-game situation.