Recently, we introduced a frequency domain measure - the power transfer function - to predict the convergence rate, system stability bound and the steady-state behavior across time and frequency of a least mean square based feedback/echo cancellation algorithm in a general multiple-microphone and single-loudspeaker system. In this work, we extend the theoretical analysis to the normalized least mean square and recursive least squares algorithms. Furthermore, we compare and discuss the system behaviors in terms of the power transfer function for all three adaptive algorithms.