You still need to have one IP address for the cluster in each network you want it to handle, unless you intend to run in bridge mode.
Each cluster member also needs its own unique IP address in the same network as all other cluster members. The members send heartbeats to each other using these addresses. These per-member IPs don't need to be in the same networks you plan to actually use for traffic. You need to use a separate network per interface. That is, you can't use 10.20.30.1/24 for eth1, 10.20.30.2/24 for eth2, and so on.
Configuring VMAC is pretty easy. It's a checkbox in the cluster settings (same place where you pick between HA and load-sharing clustering modes). When you can take an outage, check the box and push policy. The cluster members will start using the virtual MAC. You do need an outage window, because other endpoints on the networks may not see the MAC change until their ARP entries time out.