usually there is not much to do on MWG. It is built to keep itself running to provide a good availability.
What I would recommend is running the policy check indicated in the first post once you finished your policy. There is no need to regularly check the policy unless changes were made.
You should ensure that you enable log file rotation/deletion to make sure the disk is not filled with data. It doesn't hurt to check the free disk space once in a while or add an automated check for free disk space as MWG will get into trouble when it runs out of space. If log rotation/deletion is enabled and you have not manually enabled any debug options there is nothing to cleanup.
What I personally would do is enabling SNMP on all boxes and use a tool that can grab the SNMP values for CPU, memory etc from every box and write it into its database. You probably have an enterprise-ready tool like this for monitoring different network devices or use an open source tool like Cacti or similar. Such a tool allows you to create "long-term" graphs of many performance indicators in MWG, so you could check them every month and make sure they look good. This will also help you to notice when probably the number of requests per second increased over time and the sizing may cause problems or if can help you (depending on what values you grab) to notice network outages or similar problems.