This Spatial Model replaces NoData locations with the value of their nearest (non-NoData) neighbor. This is useful for filling areas of a raster where the data is known to be erroneous.
Sometimes the data which occupies a raster you produce may be sparse. I.e., the class pixels may only occupy a very limited set of locations in the image, but the extent of the image is still set to that of the original image being analysed. This Spatial Model automatically identifies the smallest bounding rectangle needed to subset out the data of interest.