But as it stands, I don't see any testing of those drivers, and by nature, I'm skeptical of everything. If I were going to spend alot on drivers right now, regardless of my liking for JBL, it would be on a pair of Seas Excel mids with cheap seas aluminum tweeters. (which test VERY well) This would put me in the $500 ballpark. Not bad in my opinion for a set of some of the best drivers in the world.
I agree with you, though I'd never personally use anything but widebanders or coincident/concentric drivers in a nearfield listening environment, home or car.
Besides, the $800 ones are a statement product, with money spent on factors having nothing to do with sound quality such as cosmetics and brand-name crossover parts. (And the Revel-style waveguides.)
While I've not heard them side-to-side, I would bet that the main substantive difference between them and the regular C608GTi MKII (when used as I would use either, as a coax) is that JBL keeps tighter tolerances on the $800 ones than on the ~$350 ones (going by a quick Google price search - I bought my pair for ~$180 at a Tweeter bankruptcy sale, and I'll probably end up using them in home theater rear surrounds at some point). Maybe the crossover design is different (and better), too, but honestly I think the C608GTi would be at least competitive with the driver I wish Seas made (Excel motor, "G" cone, Uni-Q style coax with the H1333's current tweeter), let alone anything currently on offer from them, for most car installs.
I wouldn't be going into this much detail, except that I feel like I criticize the largely pathetic and reactionary efforts of the car-fi industry so much that it's only fair when a company produces something that's actually both good and worth the asking price I go out of my way to praise said product and the company that produced it.