I don’t know if it’s ideal, but when I need things to work like that, I usually join them all into a single object to unwrap them together, apply an appropriate scale to the whole thing (like if I’m going to be using 4 textures, s 4 enter), then separate the objects and place each object’s islands into UV bounds.
When doing something like polygonal hair, where I want it to have well organized quads for anisotropy or detail normal mapping, I’ll start with a lightmap project on the entire object to get uniform scale and then follow active quads on separate strands.
Note that in a lot of material workflows, the remapping process is unnecessary; there’s rarely a difference between using a single 4096^2 texture and using 4 2048^2 textures. The former can be preferable (fewer draw calls in something like Unity, less wasted space from islands) but some apps or maybe cards bug out on really big textures.