Some key work on the Paleo diet was performed by Eaton and Konner (1997). They claimed that the Paleolithic diet would have included 37% of energy from protein, 41% from carbs, and 22% from fat, along wtih vitamins, minerals, and fiber ingested at rates that would be unachievalbe via modern diets. Nestle (2000) expresses skepticism about Eaton and Konner's conclusions, claiming that we don't know much about the Paleo diet and that data suggests that a diet rich in veggies, fruits, and grains may, in fact, be optimal.
Frassetto and others (2009) have a study that suggests rapid benefits for the Paleo diet relative to a normal diet. Since the normal American diet's so bad, it would be nice to have some more meaningful comparisons. I also wonder how much sense it makes to have honey as part of the diet if its goal is to mimick ancient diets. My guess is that many cavemen walked around their entire lives without eating any honey or sweet (by our standards) fruits. Most of the proven diets actually have a great deal in common. The Paleo diet probably has far too much meat. But if this is remedied, the idea of eating veggies and protein rich foods with some nuts and no dairy products is probably common to all of the really good diets.
Hockett and Haws (2003) have a paper on "nutritional ecology" arguing that greater diversity within the paleolithic diet led to advantages in terms of longevity and infant survival. The divide the diet into mammal meat, organs, shellfish, fish, and plants.