Dual-fuel combustion is an attractive approach for utilising alternative fuels such as natural gas in compression-ignition internal combustion engines. In this concept, a more reactive fuel is injected in order to provide a source of ignition for the premixed natural gas/air, combining the high efficiency of a compression-ignition engine with the relatively low emissions associated with natural gas. The flame modes present in dual-fuel engines impose a challenge for existing turbulent combustion models. Following ignition, flame propagates through a partially-reacted and inhomogeneous mixture of the two fuels. The objective of this study is to test a new modelling approach that combines the ability of the Conditional Moment Closure (CMC) approach to describe autoignition of fuel sprays with the ability of the G-equation approach to describe the subsequent flame propagation. The combined approach is formulated in order to ensure thermodynamic consistency in the coupling of the two models, with both fuels contributing to the heat release. This methodology can be used for the full range of fuel substitution from perfectly-premixed through to pure diesel operation. The hybrid model is coupled with the flow field computed with Star-CD software in order to simulate n-heptane pilot jet-ignited combustion of a premixed methane air charge in a rapid compression-expansion machine apparatus. The results show the impact of a new laminar flame speed model that accounts for effects of the pre-ignition chemistry in the dual-fuel mixture and the importance of wall heat transfer modelling on prediction of the cumulative heat release. The results show that the hybrid model adequately captures not only ignition and transition to premixed flame propagation, but that the CMC also captures a third stage of the combustion process in which the premixed end-gas autoignites.