While it is important to understand what information a histogram is giving you, it is also important to understand what a histogram is not telling you. Remember that each column in the histogram represents how many pixels in the photograph have the pixel value represented by the column. However, the histogram does not tell you where those pixels are located within the image. As a result, two different images can result in the same histogram.
Figure 1 illustrates a case where two different digital images can result in the exact same histogram. On the left, image 1 is divided into two halves, one with 25% gray and one with 75% gray. On the right, image 2 is divided into four quarters and colored in a checkerboard-style manner. Two of the quarters are colored with 25% gray and two with 75% gray. Both images are the same size (i.e. have the same total number of pixels). Both images have the same amount of area colored with 25% gray and the same amount of area colored with 75% gray.
The resulting histogram will for both of these images will be exactly the same. Only two pixel values are present in the images, the pixel value associated with 25% gray (63) and with 75% gray (191). As a result, only these two columns in the histogram will have any pixels in them. All other columns will have zero height (i.e. are empty). Because there are an equal number of pixels in both images with these pixel values, 25% and 75% gray, these two columns will be of equal height, as shown in figure 1. From this figure, we can see that a histogram does not communicate the structures and/or patterns present within the associated image. It only gives the viewer an idea of the distribution of pixel values.
When you are applying manipulations on an image, you should keep this idea in mind. For example, if you decide there are too many dark pixels in an image, you may want brighten up the dark areas of the image (essentially what "fill light" does in Camera Raw). However, some of those dark pixels might be next to medium gray pixels while other might be next to bright white pixels. As those dark pixels are made brighter (i.e. the pixel value is increased), they are getting closer to the gray pixels and thus the contrast between those pixels in the image is being decreased. However, while the contrast between the dark pixels and the white pixels is also being decreased, this contrast remains relatively high. The result is that parts of the image might be losing some contrast relative to other portions of the image.