What is the difference between Local/Remote/Dynamic SSH tunneling?

From the puTTY documentation, specifically, 4.23 The Tunnels Panel section:

Set one of the ‘Local’ or ‘Remote’ radio buttons, depending on whether you want to forward a local port to a remote destination (‘Local’) or forward a remote port to a local destination (‘Remote’). Alternatively, select ‘Dynamic’ if you want PuTTY to provide a local SOCKS 4/4A/5 proxy on a local port (note that this proxy only supports TCP connections; the SSH protocol does not support forwarding UDP).

• Local -- Forward local port to remote host.
• Remote -- Forward remote port to local host.
• Dynamic -- Act as a SOCKS proxy. This requires special support from the software that connects to it, however the destination address is obtained dynamically at runtime rather than being fixed in advance.

The answer from jscott is correct, however after reading it, it was still not completely clear to me when should I use local and when remote. So I researched further, and I have found the answer here:

Use local if you have a service running on a machine that can be reached from the remote machine, and you want to access it directly from the local machine. After setting up the tunneling you will be able to access the service using your local host IP (127.0.0.1)

Use remote if you have a service that can be reached from the local machine, and you need to make it available to the remote machine. It opens the listening socket on the machine you have used SSH to log into.

Dynamic is like local, but on the client side it behaves like a SOCKS proxy. Use it if you need to connect with a software that expects SOCKS forwarding.