This work investigates the benefit of using multiple signals and preprocessing strategies for sleep staging from cardiorespiratory signals. We modified our previous Neural Network model to take different signal combinations as input. To that end, we added oxygen saturation and different respiratory signals to the electrocardiogram. We further invoked different preprocessing strategies that have been described previously for such signals, namely using downsampled signals vs. using time series of breath-to-breath intervals. We trained and tested our model variations with 4784 polysomnograms from the Sleep Heart Health Study. We found the best combination of signals to be heart rate together with a downsampled respiratory signal. The classification resulted in a k of 0.68 on hold-out test data, which outperforms our previous results and state of the art for cardiorespiratory sleep staging. We observe that combinations of cardiorespiratory signals can improve classification performance for automatic cardiorespiratory sleep staging. As there are generally more cardiorespiratory signals available and many more options for preprocessing them, we expect that further research in this area will show even more improvements.