Monte Carlo-based PET simulators are powerful tools in the evaluation and validation of new PET algorithms. Accurate generation of projection data from spatiotemporal tracer distributions enable, for a given scanner specification and attenuating media distribution, quantitative analysis based on known ground truth. High activity-related phenomena, such as the contribution of randoms, as well as block and system deadtimes, corrupt actual PET scan data and therefore must be integrated within the simulation model, along with photon interactions within tissue and scanner materials. The PET-SORTEO Monte Carlo simulator, dedicated to full ring tomographs, is able to generate scattered, unscattered, and randoms event distributions from voxelized phantoms, accounting for data losses due to system deadtime. We show the results of extending the simulator to include accurate generation of list-mode data. Our implementation avoids incorrect event distribution and event timing inaccuracies cause by local and propagating temporal rounding errors. List-mode events produced by the PET-SORTEO simulator, when rebinned, are now consistent with sinograms produced by the simulator.