Jumping in here real quick...
I'm pretty sure this is not something exposed in the ZAC at this time. I've actually never seen it myself until just now. 
If it's in the ZAC, I don't realize it. @rgalletto - there's no ZAC page for "identity-configs" yet, is there?
I don't think you need to worry about addressable terminators just yet. Addressable terminators allow you to basically connect to a specific identity. Your question here, assuming I follow along, is more "how do i make sure the right identity binds the service". That you control through service policies. You simply inform the overlay using service bind policies that an identity should host/bind a specific service and it'll take care of the rest for you. I think you should/could do this with just a regular bind service policy.
Addressable terminators allow you to effectively specify the "name" of a specific termination point. Think about ssh. For ssh, you don't want to have to create thousands of services, each with a bespoke intercept address. Instead, you want to be able to target one specific destination or terminator. With OpenZiti, the identity can inform the overlay what that address or unique name is. So when my local tunneler spins up, it can tell the overlay, "bind the ssh service, but add this extra bit of information, the instanceId, to the terminator and call it clint-ssh". Then when yours comes online, you too can bind the ssh service, but call it "nick-ssh". At this point, there are two terminators in the overlay: clint-ssh, nick-ssh. This allows a client to choose exactly which identity (or terminator) to target. Dial "clint-ssh" or "nick-ssh".
Now when you use bindUsingEdgeIdentity with tunnelers, that's just a shortcut to the configuration saying "whatever my name is, use that name as the terminator instanceId". You could also set it yourself if you prefer. This does mean that I could advertise "nick-ssh" from my identity. That's a different problem we'll be looking to correct sometime in the future.
Hopefully that extra context helps?