I love your little sprite! And actually, the bigger one helped me a lot to understand what you're going for, so I could give meaningful further help (hopefully) with the tiny one. I hope you don't mind, I did a small edit mainly to convey a pretty simple but on the face of it counterintuitive principle. Here's the edit:

What I wanted to tell you is that sometimes to make something more readable, you have to actually make it chunkier and to take detail away. You've squeezed a lot of character in this sprite, but I think you can go a bit further with tightening up the shapes that you do use, and (this is always my broken-record advice) consider how many single pixels you actually need, as opposed to larger clusters. Their arrangement conveys depth and priority better, I feel, if they're not just, you know two pixels in a corner in a diagonal arrangement, ambiguous and abstract for the mind to solve. Connect them in a tetromino, maximize your usage of a shape, if you're going to imply it anyway.
Your rendering is also quite soft, which I don't mind, it depends on what sort of background it'll go on, but I thought a bit more contrast might be worth exploring. You've got a looooaaadd of near-colors that generally in pixel art will only get in your way when you go to animate. I didn't clean the palette too much, but I just wanted to show you some cluster theory, because it'll help you so much to animate things later if you know why and how each of your shapes exists. Take the little animation in your pixel art program and flip between the two frames and a lot of changes I made will become apparent in their function. it's like a cascade, you fix one little cluster, then everything around that will have to shift to accommodate. Does that make sense?