There are two main types of image stacking, one for essentially a static field, with multiple images of the same thing being averaged to reduce noise and improve the signal-to-noise ratio. This can be done in Image Stacker (The manual, with examples, can be found here).
First, I show a four image stack of a diatom with a 60x Plan Apo objective used with circular oblique illumination, then a comparison of one of the frames vs. the stack, showing the noise reduction capabilities of the Image Stacker software.
Notice how much more granular the top of the last image is compared to the bottom. This is done by stacking (averaging), not Photoshop noise reduction of any sort. Only auto-contrast, crop and sizing is done in Photoshop.
Where Image Stacker falls short is with focus stacks, particularly if they are deep or bright, for example, some fresh moss taken with Disco Lights Illumination (epi-Rheinberg) has far too much depth and out-of-focus information for simple stacking to deal with. Shown here is an animated .GIF (10.3 MB .GIF) going through focus, and the best result obtained with Image Stacker averaging (23 images).
The other type is more complicated, focus stacking, where the narrow band of in-focus field is different for different focal positions, and the software, Zerene Stacker, being one example, uses the in-focus parts of the image to recreate the full field in focus simultaneously.
A further advantage of Zerene Stacker is that it "knows" about the specimen in 3-D, so can create stereo pairs and animations of "rocking" the specimen back and forth. For example, here is the moss again (7.3 MB .GIF).