Oh, is that why models are often MASSIVE and you have to scale down in Blitz?
No, you misunderstand what he was saying. He wasn't talking about the size of the model, he was talking about the UV coordinates. The size of the UV coordinates has no relation to the size of the model. When the polygons are really small on the UV map, that means only a few pixels are being stretched over that polygon. Thus, obviously the lines will look ugly.
Meanwhile, in your other model where it looks good, there aren't as many polygons so each is larger in the UV map and gets more pixels on the texture.
The main issue is that you are wasting a lot of space (on the texture) by having each polygon mapped to it's own unique spot. This is the default for the mapping you are doing, but you should then adjust the coordinates so that polygons are overlapping. Specifically, if two polygons will have the exact same coloration, then you can overlap the UV coordinates and just paint it once. By overlapping polygons, you'll save space on the texture and can make things bigger.
Referring to your first screenshot, see how there are all those tiny polygons in UU? Well, you can overlap all the polygons that are the same color, making sure to tweak the UVs so that all the vertices line up, resulting in what looks like a single polygon but is in fact a whole bunch overlapping. Then you scale that up to give it more space on the texture (this is what 3D artists are referring to when they talk about "texture budget.")
As for these questions:
Just how the hell does any one texture things correctly if the seams never work correctly.
How the hell do pros create really detailed models (with small 'bits') with only a tiny single UV map?
well yes, UV mapping is pretty difficult. Whatever gave you the impression that it was easy?
