What's the point of FortiClient Application Firewall?

I've had it managed with EMS and enabled for my endpoint profiles for years, but never gave it much thought before. Specifically, the categories that you can enable, block, or monitor. If my endpoints are behind a Fortigate with web and application filters, the application firewall policies seem redundant.

I have all the categories set to approve, because I want to control that with my FG, not EMS profiles. I do have the "Block Known Communication Channels Used by Attackers" so I guess that's some good it's doing.