Queueing network models with finite capacity queues and blocking are used to represent systems with finite capacity resource constraints, such as production, communication and computer systems. Various blocking mechanisms have been defined in the literature to represent the different behaviours of real systems with limited resources. In this paper we propose a technique that allows to represent this type of queueing networks by means of generalized stochastic Petri nets. The method allows to obtain several benefits both for the qualitative and the quantitative analysis of these queueing networks. In particular it offers the possibility of using results and tools developed within the framework of the Petri nets. In the paper some of these potentialities are presented.