You don't assign scopes to particular NICs. An IP will always be pulled from the pool that is within the subnet requesting the IP.
For example. Say your server is connected to 192.168.1.x and 192.168.2.x.
You setup 2 scopes, one for each subnet; say 192.168.1.50 to 192.168.1.99 and 192.168.2.50 to 192.168.2.99
When a client from the 1.x network requests and IP it will get an address from the scope where IPs are in the 192.168.1.50 to 192.168.1.99 range. The same idea applies for the second subnet.
Also, if you setup a third scope, say 192.168.3.50 to 192.168.3.99; and a DHCP Relay agent who has a NIC connected to a 192.168.3.x network can then forward those requests to the server.