Help understanding routing in site-to-site-to-site VPN

For a few years now, I've been trying to set up this system to allow me remote access to networks.

At the remote network I want to put a cheap little router, I got some Next WT3020 2 ethernet port mini routers and flashed them with OpenWrt.
That router will establish a VPN connection to a server.
I can then establish another VPN connection to that server from my computer and gain access to the network the little router is on.

I tried a while back with OpenVpn, and I never got it right.

I've currently got an Linux cloud hosted VM running Wireguard 'server' (know they're all nodes)
I can establish a VPN connection into it from both ends. and the little router can ping the 'server' and the my computer's end of the Wireguard but I can't see into the 'far' network from my computer.

when my computer connects to the Wireguard server I'm pretty sure its using Wireguard for all traffic not on its local subnet. I think "AllowedIPs 0.0.0.0/0" in the Wireguard config does this.

The Openwrt router is basically stock with just the config for the, Wireguard connection entered and I've tried adding the Wireguard connection to the LAN Firewall zone.

The WG config on the 'server' was generated buy a shell script I pulled off GIT.

Any advice on how to proceed would be very welcome, I've been at this for years, having a go then abandoning it through frustration or lack of spare time, but right now it would be super useful what with the world being in lockdown.

If you're wondering what its for, I build control panels for industrial applications and I often need to make software changes to them. There are some off the shelf solutions for this, which are all great, some of them even run on OpenWRT on their routers, but they're super expensive and despite all my efforts I'm seldom able to convince people to buy them.