Ejector refrigeration cycles offer an alternative to traditional systems for the production of cooling using low temperature heat. In this paper, a real gas thermodynamic model based on the mass flow rate maximisation is presented. This model has the advantage of simplifying the calculation algorithm and avoiding a complex description of the double choking mechanism taking place within the ejector. First, the model hypothesis and calculation algorithm are presented. The impact of each efficiency is evaluated and a tuning procedure is developed to calibrate the model on experimental data. Validation is performed on multiple datasets relative to two different fluids: R600a and R134a. The ejector model is then used to simulate a SERS (single ejector refrigeration system) cycle, to validate its robustness and capability to be used in the prediction of thermodynamic cycles performance. A comparative analysis of different fluids is carried out on the SERS, highlighting the important role played by the choice of the superheating. Finally, the model is used to predict performance in the case of a two-phase primary flow pointing out the limits of the model and the need of further experimental studies for the inclusion of appropriate semi-empirical corrections.