Simply articulating your questions often focuses your attention well enough to reveal a good answer without even making a paint and paper study.
I think the COLOR is fine as it is. Besides, changing the colors won't eliminate any detail.
How about EDGE QUALITY?
Bingo! I can make as many blades of grass and leaves as I please. As long as they have soft edges they will feel like part of a single larger shape.
Let's try another image:
Here's one with a definite foreground, middle ground, background composition, which should be fun, except the upper left middleground is trying to push forward into the foreground. What is it about that area that makes it so assertive? Let's zoom in...
Going down the list to find the culprit in the sand dune picture never got to the fourth variable, VALUE . Be sure to include that one when you analyze this scene.
Start by stating the question you want to answer as simply as possible. Then, see if a possible solution arises from your inquiry. If so, onward you go! If not, go down the list.
If you can picture the changes you want to make, and you're confident that they will solve the problem, you may not need to make an actual study. If you're not sure you've got the answer, get a small piece of paper and try out your idea. Remember, a study is not meant to be frameable. Keep it very simple and quick.
For homework, choose an image, then ask yourself, "What looks tricky?"
Write down your observations in the form of a question that might begin with, "How can I...?", or, "What can I adjust...?"
Write down what your analysis reveals.
Design a study that will provide the answer to any questions that linger.
Paint the study.
The assumption here is that it's OK to deliberately diverge from accuracy. Are you onboard with this? Personally, I think that's our job. There's a reason we call it "art", as in artifice.