In this study, we applied the Total Monte Carlo (TMC) methodology in de-excitation simulations of primary fission fragments (FF) using the TALYS code. The goal was to develop and optimise a methodology to benchmark initial fission model assumptions on fission mass yield distributions, excitation energy sharing and angular momentum population. The study was performed on the thermal neutron induced fission of Pu-239(n(th),f). The work aimed at evaluating fission model deficiencies and parameter sensitivities. We systematically varied TALYS input data by generating 5000 random files through the GEF code, randomizing 94 model parameters that affect fission yields and energy distributions within 3% of their default values. This variation revealed significant changes in the fission observables, such as prompt neutron and gamma-ray multiplicities and energy spectra. The results indicate some systematic defects in the assumed excitation-energies and angular momenta. Another outcome from the study is the identification of a need for new correlation measurements on prompt neutrons and gamma-rays from the Pu-239(n(th),f) reaction, as well as an updated evaluation.