They should both be hyphenated, "First-Person" (FP or 1P) and "Third-Person" (3P). The terms come from the language perspectives of first person, "I" and "we", the second person, "you" and "you all", and the third person, "he, she, it" and "they".
When you add a completely different item to be held, the tags you must develop are:
- first-person model(s) of the object
- third-person model(s) of the object (this can be a duplicate of the fp model(s), or it can be a lower-quality model with fewer polygons)
The detailed objects, those with more polygons, are the objects that are closer to the camera, the FP models. The same principle applies to Levels of Detail (LODs).
- first-person animations of the biped's arms holding the object (what you see in first-person view)
- third-person animations of the biped's body in several engine-supported movements, like crouching, jumping, turning, running, and getting hit or stunned from four different directions and at different locations on the body
The third-person animations are typically reused because the Halo engine is only supporting 256 or 257 (but that's the breaking point) animations per model_animations tag. This is why, for example, the flamethrower uses the sniper right reloading animation in third person.
The first-person and third-person things don't have to match; what you see from your perspective and what you see the other cyborg model doing does not have to be the same; neither the models nor the animations must appear the same between perspectives. Each perspective's model and its animations must have the same node count, though, which is like a checksum for telling the engine that the animations match that model and so the model will animate.
The rest will be in the video tutorial when it's redone.
Either you are groping for answers, or you are asking God and listening to Jesus.