IP assignment strategy with UnRaid server

I'm planning a server that will run UnRaid as a NAS providing a shared file system to Windows Server and Debian Linux VMs. 

I want the file system (user shares) from the parity-protected array to be available to the Linux and Windows VMs, as well as to other users over the LAN.

From this documentation---

Assuming I create the "public bridge" for the two VMs, then I would see the Windows server at it's own IP, as well as the Linux server at its own IP, in addition to the IP of UnRaid's file system.

So from another system on the network, I could see these hosts:

192.168.1.1    UnRaid NAS with hostname fileserver      \\fileserver\share1,  \\fileserver\share2, etc

192.168.1.2   Windows Server VM -    Used to access windows server with Remote Desktop, etc.

192.168.1.3   Linux Server VM -    Used to access Linux server with VNC etc.

The Windows and Linux can mount shares just like any other device, at \\fileserver\share1,  \\fileserver\share2, etc.

Windows and Linux boot off of a separate SSD drive (or 2 drives) that are their configured a cache-only share - not part of the large Parity-protected array (built with rotating drives).  These drives would not be shared on the network.

So in terms of drives, the UnRaid system would have:

USB Boot Drive for UnRaid itself

NVme SSDs for VM boot drives (cache only)  (Question - do these need redundancy too? A total of 4, two each for each VM, or two SSD total, each with two partitions, one partiion for each VM)?

Parity-protected array (5 or 6 large drives for NAS)