Again, to learn a complete model of something you need both grid cells and place cells. Grid cells create a reference frame to specify locations and plan movements. But you also need sensed information, represented by place cells, to associate sensory input with locations in the reference frame.
