The way I see is that $1...$n are related to the inheritance concept. They could be used inside [send~] & [receive~] objects to force some sort of locality, but you can't really guarantee locality by that, it is just some way around that is not 100% safe

There's no guarantee of "locality" whatsoever in using $0 either. If you create a [send $0-xxx] and the identifier of the containing (instance of an) abstraction happens to be 1032, and then in some other patch or abstraction you have a literal [receive 1032-xxx], the "locality" will be lost.

Using $0 as a prefix to the name of "local" sends and receives preserves locality only provided that you never use "literal" numbers as a prefix in the name of sends or receives.

So $0 is not so different from other creation arguments (i.e. $1...); the difference is that it is assigned automatically by the system and that it is done in such a way that no two instances of any abstraction will have the same value of $0.

Using it for creating "local" names is just one way of using it, just as you can use explicit creation arguments for the same purpose (by taking the responsibility of ensuring uniqueness), although it is probably its main "intended" use.