Yes, I think this would be useful for port-knocking, or when detecting that a system outside has successfully established a connection (logged in), it can be removed from the blacklist.
I do think it can be worked aorund by instead adding the address to a whitelist (and then possibly having a script come around every x minutes checking the whitelists and deleting the whitelisted entries from the blacklists). But that would be much more complicated and likely error prone