And yet I as a very competent gamer and DM as I am sure you are (No snarkiness) see 4e as an extreme deviation from what D&D previously was. I see differences in gameplay, philosophy, and construction.
I really do not get the assumption of PC's fighting one monster or monster type at a time. I never delivered encounters that way.
I also never accepted rules from the Book fo NIne swords. I did not like the direction of D&D then. I was more disappointed with 4e. I was hoping for a D&D SAGA release.
In any case, i see 4e as an extreme variation off of the previous design models.
I'll both agree and disagree on various points. I do agree that 4e is a strong variation over the designs of 3e, in many regards, design philosophy being chiefly among them.
3e used a cohesive, integrated design. All the parts worked together to form a giant clock of a game, which the more you get into it, the more you see. The problem wound up being that integrated design, while very strong by itself, does not lend itself to expansion very well. Take a look at the Tome of Magic and Book of Nine Swords, both of which had very mixed responses, because they took those integrated mechanics and tried to fundamentally alter them. Same with psionics, and any number of splatbooks. IMHO, 3.5 was at it's best with a small selection of books available, and even then mostly drawing from the core three.
4e, by contrast, uses a modular design. Each individual piece is designed to work cohesively within itself, without altering the game as a whole, or anything else in the game. You never have to worry about Class X taking a Prestige Class designed for Class Y and having unintended effect Z, which shoves Character Alpha far above or below the power curve. Yes, you lose some customization in the process, but the modular approach makes expansion much easier, and IMHO delays true bloat.
When 3.5 came out, I bought the three core books, and proceeded to buy books for about a year and a half. After that, there was such a drop in quality, and things got so difficult for me to keep track of, I stopped buying books. Now 4e is over two years into it's life, and I'm still buying most all the hardcovers they release, because I feel they will enhance my game experience, and indeed they do.
3e was the beginning, and at least for now the end, of integrated game design for D&D. While second edition could have been argued to have a more modular approach, there are times where I doubt these ideas were really thought out at that time. Behold the advancement of game design over the years. The jump from 3e to 4e keeps the framework of the system. You still roll a d20, add modifiers, compare them to a difficulty and determine success. Rolling high is always good. The action economy is basically the same, if a little more solidly defined, tactical positioning is important and use of a battlemat is highly recommended. Opportunity attacks remain, as does Vancian casting, just with a few more restrictions and applied across the spectrum. Class design, indeed what class even means, is different, as is monster design. Both these resemble 2e far more than 3e. When 2e jumped to 3e on the other hand, your peripherals of the game (classes, monsters, etc.) completely changed their definition, as did the majority of the system. Complicated math made way to streamlined math, the battlemat went from being completely optional to highly reccomended, the action economy changed as well as many of the baseline mechanics. D&D went from a hodgepodge of rules built around a single core to a large, humming, efficient machine.
Both the 2e to 3e and the 3e to 4e transitions were big steps. I see the 2e to 3e step as much larger, at least 3e and 4e share the same basic mechanical guts. Every edition so far of D&D has been a huge change, though the reasons can be debated until the end of the roleplaying hobby. I feel it's due to the complexity of the system, but that's just me.