The 24mp replacement for the D700 rumors have returned (see Nikon Rumors). I know Nikon did work on such a camera, but the lack of video in the original Sony sensor and the 5DII's video success may have scuttled it.
I still think Nikon has a tough choice with the D700 update: a D700s makes a lot of sense, and there's no other sensor that does what that D3s 12mp sensor does--we'd all hate to see it die away. But as much as many disclaim it, the crowd demands megapixels, so a 24mp update also makes sense, especially if it can improve on the D3x and add video. It's one of those damned if you do, damned if you don't quandaries that come up with product progressions every once in awhile.
Some might say "just do both," but I think that's a bit of a stretch for Nikon. Three FX bodies have pretty much pushed them to the limits as it is. Adding more models isn't going to make supply/demand issues any better, and it also introduces all sorts of marketing problems that Nikon hasn't historically been good at getting under control.
So I think Nikon will do a single D700 replacement and probably initially a single D3 series replacement. Call them the D800 and D4. As I've written before, putting the D800 at 24mp and the D4 at fewer megapixels introduces all sorts of problems of its own: the D4 at, say, 16mp, would have to perform like a D3s or better at high ISO values, and that's a bit of a stretch to imagine sensor-wise. But putting the D800 at 12mp with the old D3s sensor and the D4 at 16mp isn't all that much better.
One of the problems is that we're in the land of diminishing returns now. Doubling sensor pixel count (12mp to 24mp) is not a doubling of resolution. It increases resolution 40%. Most studies have shown that the majority can't see increases of less than 20%, so 40% is a modest visual impact, not the "double" that the numbers imply. And don't forget to throw in diffraction and noise impacts, which will whittle a bit more off the end result.
On the other hand, lower noise levels are relatively easy for most people to see, and you don't need to "double" noise reduction to have high visual impacts. It's one of the reasons why the D3s shines. Unfortunately, while sensors continue to move forward in terms of noise (and/or dynamic range), the D3s was already above the expected progression there, so the prospect of a 24mp sensor with D3s-like noise properties is well off the prediction curve.
There's no easy answer here. A D4 is a given, as Nikon uses the new pro bodies to introduce new technologies into the DSLR line. The question is what Nikon will do in the gap between the D7000 and the D4. Right now, two cameras sit in that gap, the D300s and D700, and both need updating to remain competitive. For the high-end user, 2011 is going to be an interesting year, as it could produce three pro bodies (D400, D800, D4). We're all waiting to see how Nikon sees the differentiation between these products.