Yeah, all voltage sources can be redrawn as short circuits, and all current sources can be redrawn as open circuits, for purposes of dynamics.
This is because, for a given perturbation of the system, only a frequency at the same frequency as the source could have any interaction with it; for all other frequencies, it must act as a short circuit. Unequal frequencies are orthogonal, so there's nothing to equate with, no way to interfere.
You can imagine, if a stimulus were applied to the circuit some other way (say through a high value resistor, or a magnetically induced current*), then the circuit must respond the same way (i.e., it exhibits the same dynamic equations).
It's only a small stretch further to say: well, if it obeys the same dynamics, and the circuit obeys superposition (it does, it's linear), then it must also be true for equal frequency (and phase) as well; and so, the circuit will respond that way to the original source exactly the same as it does to another stimulus.
*Which, mind that these are simply other kinds of sources in the equivalent circuit, and therefore are still within this rule. I can't make a non-circular argument here, because if it's a LTI source, it must fit into the theorem I'm trying to explain...
Or, to put it more directly: if the superposition theorem applies (i.e., you can solve for the V, I in a circuit by setting all but one independent sources to zero, cycling through each one at a time, recording the node/branch V/I conditions from each, then adding them all up), then the circuit must have the same dynamics whether the source is zero or nonzero.
Tim