Starting help with linear control theory

Hey, was wondering if perhaps someone here had time to check my thinking and let me know if I'm horribly off the mark.

So we have a really simple controller (I'm attaching a block and Matlab Simulink diagram) with no feedback and want to write a transfer function. T stands for torque in this case; one part being disturbance and the other control. As I understand it I simply start on the right and see that

theta = (1/s)*(dtheta/dt),

with s being the Laplace variable. Then I just work my way to the left in the same manner, ending up with (T being the input)

theta = T*(1/I)*(1/s^2)*e^(-cs).

Looking at it from a physics standpoint it seems right to me. Torque is angular acceleration times inertia; divide out the inertia and integrate two times to get an angle resulting from the input torque. The step function then modifies the input. I'm not sure I've understood it's purpose entirely, but I think it has to with the system having a small measure of friction so there won't be any response before a minimum torque is applied. Anyhows, I'm completely new to this and really was just wondering if I am thinking about this in the right way?