Oy vey!
Dreams....oh we love to have em explained. Explanations are typically freudian inferences with no science backing it up.
I adhere to the restoration theory.
Our memory functions by having the neural networks associated with a memory an object/situation regularly activated.
This is the 'use it or lose it' principle and is scientifically sound. If certain neural networks aren't activated the connections get weaker, less likely to be activated both directly or by association. When we sleep our brain's activity level is still around 90%, the restoration hypothesis argues that the brain randomly fires neurons to keep them intact and preserve the associated memories. Recently activated networks have an increased chance of getting 'sparked' in this proces because thye have strong and many links to other networks. You end up with recent issues that have bothered you or are important, mixed with random elements coming up during our dreams. Nightmares are a good example of the 'use it or lose it' principle:
a bad dream can scare you, you think about it, strenghtening the network and it's connections increasing the chance it will pop up in the next dream. This is a vicious cycle and very hard to escape.
This is a very parsimonious and elegant theory, based on proven neural principals. I'll take this over new-age or freudian inferences which are always biased by zeitgeist, culture and personality.