I can see the remote displacement is scoped to the top face and blend radius of the countersunk hole.
Since a tensioned bolt clamps the flange to a frame not modeled, a common approach is to apply the remote displacement (behavior Deformable) to all the nodes in a cone or cylinder below the head OD. It would be ideal to go back to SpaceClaim, and slice that solid from the bracket and reconnect it using the Share button to make it simple to scope the remote displacement to that tube-like body. The benefit of that is that the ring around each bolt hole on the bottom of the bracket that is clamped to the frame (not modeled) is constrained from moving due to the friction of the interface and the bolt tension. If you don't want to expend that effort, an approximation to that is to scope the remote displacement to the ID of the hole as well as the top face of the countersink. I would not include the blend radius.
I would set all six DOF of each remote displacement to 0.
I see you plan to use a bearing load on the 0.75" pin bearing faces. That is a good way to distribute the forces. For the applied torque, I would decompose that into the normal forces which would be equal and opposite on the two bearing faces.
For mesh refinement, evaluate all load cases.