This is not true.  The estimate of the overall dynamic model by this method would not be accurate.  For higher order systems, the output is not well approximated by a first order response.  An apparent dead time occurs that would not be accounted for by the estimation proposed in this question. See Figure 5.5 for sample dynamic responses.

Note that the expression for the overall steady-state gain is correct.