They can both be important, but I think that in most cases, content, at least in the sense that we’re talking about here, is more important than art and graphics, although it kind of depends on what kind of game you’re making. (puzzle games and RPGs for example, traditionally benefit much more from content than visual art, while a multiplayer fighting game is pretty much the reverse) Personally, I would also classify a lot of game mechanics in a different and even more fundamental category than content or art, but that’s just how I tend to think about it.
The way I see it, higher quality content (good mechanics, good story, memorable characters, good level design, well designed puzzles, etc.) will make the game more fun, and the more of this content you add, the longer the game will last, further increasing its value. (again, depending on the genre) On the other hand, spending more of your time on art, either in making really high quality art or lots of individual art assets, will help your game get noticed, but it won’t necessarily make the game any longer, or any deeper, or any more fun to play.
(Since art is content, I'm using the definition of content = maps/levels/areas, art = music/textures/3Dmodels. This is a bad definition, because that means new enemies = art rather than content, which is wrong, but when we're starting with wrong definitions in the first place, there's not much you can do)
The way I’d read it, new enemies would only be considered entirely art if they’re nothing more than re-skinned versions of existing enemies. If we're separating the visual component from the gameplay components, then the 3d model itself = art, but the stats, weapons, abilities, and story behind an enemy would surely be considered to be a different kind of content than just “art.”