From a theory standpoint, the OpenVPN services in OpenWRT create one or more virtual networks that overlay another network or networks. A virtual private network (VPN) cannot, by definition, exist on its own. It always requires an underlying network over which it is laid.
The underlying network can be a physical cabled network (a basic LAN network), a physical wireless network (a basic WLAN network), an Internet- or Intranet-connected network provided to you by a third party (e.g. home broadband connection) or it can be another virtual network, such as an already-active OpenVPN network.
The process in using OpenVPN with OpenWRT always follows these steps:
1) Design the physical network, with LAN and maybe WLAN, and usually some sort of WAN connection as well
2) Implement the physical network using OpenWRT's configuration options: https://wiki.openwrt.org/doc/uci/network
3) Ensure that your physical network works the way you designed in 1). "ping www.google.com" for example
Once this is done, you can move to the OpenVPN side:
1) Design the virtual network that you want to achieve. This is usually the most difficult part
2) Implement the network settings required to achieve the virtual network. For this, you basically use the same tools as for the physical network configuration before
3) Configure the settings required by OpenVPN so it can operate on the virtual network that you just created
4) Test the network connectivity
The Wiki part "Configure the network on the OpenWRT router" is all about 2) above, that is, defining the virtual network settings that OpenVPN is supposed to operate on.