The venture into various unconventional hydrocarbon resources, such as heavy oil and oil sands, has been aggressively undertaken by industrial oil players to satisfy the ever-increasing world energy demand. Heavy oil captures the global attention due to the sheer size of its oil-in-place volume. However, its high-density and high-viscosity characteristics represent a great challenge in terms of recovery optimization. Reservoir modelling and simulation has been widely recognized as an important technique to evaluate recovery efficiency as a function of the reservoir conditions and the type of recovery process. In this paper, an extra heavy oil reservoir, the Oilfield Alpha, located in Venezuela is selected as the reservoir of interest. Conceptual models of extra heavy oil recovery, categorized into cold and thermal techniques, are addressed. Both of the thermal methods (Steam Flooding, Cyclic Steam Stimulation, Steam Assisted Gravity Drainage, etc.) and cold methods (Horizontal Well, Vapour Extraction, etc.) are studied and modelled using CMG STARS simulator. Performance indicators - typically the recovery factor, oil production rate, Steam-Oil-Ratio (SOR), etc. - for each model are evaluated and compared. The simulation results show that, after 20 years of production, the thermal methods have prevalently higher oil recovery factor, i.e. 7 - 37 % of OOIP compared to that of cold methods, i.e. 8 - 13 % of OOIP. In conclusion, the mechanistic models assist in strategic reservoir development planning at preliminary stage and, in turn, provide enhancement of oil recovery and productivity.