The MultiScale ThermoHydrologic Model (MSTHM) predicts thermohydrologic (TH) conditions in emplacement drifts and the adjoining host rock throughout the proposed nuclear-waste repository at Yucca Mountain. The MSTHM is a computationally efficient approach that accounts for TH processes occurring at a scale of a few tens of centimeters around individual waste packages and emplacement drifts, and for heat flow at the multi-kilometer scale at Yucca Mountain. The modeling effort presented here is an early investigation of the repository and is simulated at a lower temperature mode and with a different panel loading than the repository currently being considered for license application. We present these recent lower temperature mode MSTHM simulations that address the influence of repository-scale thermal-conductivity heterogeneity and the influence of preclosure operational factors affecting thermal-loading conditions. We can now accommodate a complex repository layout with emplacement drifts lying in non-parallel planes using a superposition process that combines results from multiple mountain-scale submodels. This development, along with other improvements to the MSTHM, enables more rigorous analyses of preclosure operational factors. These improvements include the ability to (1) predict TH conditions on a drift-by-drift basis, (2) represent sequential emplacement of waste packages along the drifts, and (3) incorporate distance- and time-dependent heat-removal efficiency associated with drift ventilation. Alternative approaches to addressing repository-scale thermal-conductivity heterogeneity are investigated. We find that only one of the four MSTHM submodel types needs to incorporate thermal-conductivity heterogeneity. For a particular repository design, we find that the most influential parameters are (1) percolation-flux distribution, (2) thermal-conductivity heterogeneity within the host-rock units, (3) the sequencing of waste-package emplacement, and (4) the duration of the preclosure ventilation period.