What is the best general structuring approach to building a robust 3d model for use in video games? (i.e. a character/vehicle controlled by the player/AI) Some of my questions in specific are:

1. When should I use multiple models? This seems useful for things like weapons, but what about vehicle wheels or destructible pieces - how far should you take this concept?
2. When should I use separate meshes within a single model?
3. Is there a 1:1 realtionship between meshes and materials that should be honored? It seems like meshes get sent to the shaders as a unit, although you could possibly break them down into primitives and send those pieces to different shaders...
4. Is it acceptable to animate a rigid model (like a mech) using a single mesh with disconnected pieces, assigning all the vertices in each piece a 100% weight to a corresponding bone?
5. Should you ever create animation channels for meshes (rather than a bone)?
6. Are there standards for how many texture files should be used with a single mesh or a model?

This topic may seem more to do with modeling than coding, but the reason I'm asking is because these details inform much of the design of the model handling code, which I would like to be powerful and flexible but elegant and clean as well.

Thanks for any advice!

1 & 2. This is very application-specific, and it's generally a trade-off between performance and realism/complexity. A general rule of thumb would probably be to not use multiple models/meshes for any given game entity unless it's required by the game. For example, don't make a separate model for the wheels unless they can actually come off vehicle, or change appearance in some way.

3. Personally i prefer using texture maps for material properties, otherwise i would probably stick with one material per mesh.

4. Assigning all vertices a 100% weight to their corresponding bone makes the animation look less realistic on organic objects (you lose the skin-twisting effect). If that's acceptable or not is entirely up to you :)

5. I might be wrong but I can't see any reason to animate actual vertices when there's a working rig animation system in place.

6. This is also application specific, no standards. In many cases it's more efficient to load textures separately from models and just have the models reference whatever textures it's using. This way one texture can be reused by many models, but that's a different story.