The user "would like to count nuclei and other stained cellular structures... On the attached image, I'm trying to threshold on the brown stained areas, but I'm not having much success. Any recommendations for proper threshold settings and counting on this type of image?"
I ran Pixcavator and it seems to work well with this kind (microscopy) of images. I had to switch to the blue channel which is crucial if you want to filter out those blue spots. The screenshot shows the settings (keep in mind that I also shrunk the image 2X for quickness sake).
Brown cells captured, blue ignored:
Other image analysis examples