In this work, I study a variation of the Vicsek model [2] in which a heterogeneity at the noise level is introduced (as proposed by Guisandez et al. [1]). In particular, I study the phase diagram of the system and I show that the latter undergoes the usual liquid-gas phase transition of the Vicsek model, instead of the continuous one that they claim to be present after a tricritical point in their work. These results are just another example of how finite-size effects strongly affect these kinds of models and can lead to wrong results as happened for the original Vicsek model.