In the electrophotographic printing process, the deposition of toner within the area of a given printer-addressable pixel is strongly influenced by the values of its neighboring pixels in the digital halftone image. This interaction between neighboring pixels is complex and nonlinear. To account for these effects, a printer model can be embedded in the halftoning algorithm before the printing process. Models that are designed to predict the effect of these factors on the printed halftone page can be used to design halftoning processes that will yield higher print quality, more consistently.In our previous work, we developed a strategy to account for the impact of a 5×5 neighborhood of pixels on the measured value of a printer-addressable pixel at the center of that neighborhood. We also examined the potential influence of a much larger neighborhood of pixels (45×45) on the central printeraddressable pixel. In the present paper, we improve the design of the test page for 45×45 pixel models to yield more accurate and more robust results with fewer pages. We create six different models to more accurately account for local neighborhood effects and the influence of a 45×45 neighborhood of pixels on the central printer-addressable pixel. These models have a variety of computational structures that allow system designers to choose the model that is best-suited to their particular application. They also offer varying degrees of accuracy. The model validation experimental results show that the best of these new models can yield a significant improvement in the accuracy of the prediction of the pixel values of the printed halftone image. With respect to prediction of mean absorptance (cross-validation), we gain over a 4× improvement in accuracy between the best of the six new models and our previous 5×5 model.