Combine two exposures
Start with the photo that should define the canvas, then choose a second exposure from the controls. Scale and position affect only the second image. Opacity controls how much of its blended result replaces the base.
Screen removes dark regions by combining inverse channel values, so a bright portrait against black can merge with clouds or foliage. Multiply does the opposite and keeps dark structure. Lighten compares the two images channel by channel and retains the higher value.
The film origin of the blends
On film, a double exposure happens because two scenes add light to the same negative, and light only accumulates. Screen mode is the digital counterpart of that additive behavior, which is why it produces the classic portrait-filled-with-landscape look. Multiply corresponds to stacking two slides on a lightbox, where each layer can only remove light.
Knowing which physical process a mode imitates makes the choice less trial-and-error: additive scenes want screen, subtractive overlays want multiply, and lighten is a selective mask that takes whichever photo wins per channel.
Example: silhouette filled with a forest
Use a side-lit or backlit portrait on a plain bright background as the base. Load a forest or skyline as the second exposure, set the blend to multiply, and scale it to about 140 so texture covers the whole head. Slide Position Y until the densest branches sit over the face, then lower opacity to around 70 so a trace of the facial features returns.
With screen mode, invert the plan: a dark background portrait as the base and a bright-sky landscape as the second image.
Prepare the source photos
High separation between subject and background gives the blend a readable shape. Crop close to the intended composition before combining, because the base image fixes the output dimensions. Transparent areas in the second exposure do not alter the result, and transparency in a PNG base survives to the export.