I often post images that are montages (an image that is created by overlaying two images together). I have a gallery called Dreamscapes - https://pbase.com/ed_k/dreamscape - (after the book of the same name by the Canadian photographer Andre' Gallant). In this gallery I show the three different montage types and how they are made.
The first type is the Orton Technique (developed in the '80s by Michael Orton, another Canadian photographer). It is illustrated in images #1 - #4. The idea is to make two images, #1 & #2 (originally done with slides) - one of which is overexposed and totally out of focus, #2 (how much over/out "depends"; experiment). The two images are overlaid (two slides in one mount) and the result, #4, is a soft ethereal glowing image. I've shown a tonal mask (#3) which is sometimes helpful in controlling the contribution of each image in the final blend. Using one image and creating the 2nd in software via Gaussian Blur or similar does not work nearly as well as making the out of focus image in camera. The out of focus shot must "blossom" (get larger) in order to given the characteristic Orton "glow" around subjects and Gaussian does not "grow" enough. Besides, it's faster to do it in camera than to add yet another computer step to your workflow.
The 2nd type is the Mirror Montage. It can be done several different ways. Also, each different technique can result in three different images since the montage is created by flipping (mirroring) the original image and overlaying it on itself (there are three possible ways to flip it). The basic technique is illustrated in images A, B, C, & D. A is the basic image and B, C & D are the results of three different flip/overlay combinations. I used two of these for my Freeman Patterson workshop assignment - "The uncanoeness of canoes" (I didn't choose the topic and, yes, that's a canoe). https://pbase.com/ed_k/canoes
The 3rd type is the Composite Montage, #5 - #7. It is made by overlaying two different images. Making this type of montage is straight forward. The trick is finding two images that work together. The example is an obvious application where the 2nd image is used to add texture to the final.