On this page you can see both the central queue of the agent controller and the queues of the agents.
The first table, which will always only have one line, shows the queues of the agent controller:
The second table, having one line per node, shows the queues on the nodes:
There are 4 possible queues:
Default queue for all kinds of miscellaneous tasks
Hypervisor queue takes on tasks related to virtual machine management
IO queue is typically used for backup/restore and long IO bound related work
Process is for monitoring related jobs only
In the Agent Controller Queues table, there is also a column Internal, this is not really a queue, but rather just information showing the number of internal communications going on between the Agent Controller and the worker process on the agents, mostly communication about the queues itself.
Clicking the Details link in the last column gets you to the Job Details page, showing details of the jobs