I want to preface this with the mention that understanding other people's code and being able to modify it in a way that gets it to do what you want is a big part of real world coding and not a small feat.
The rest of my comment may come across as "you're learning wrong". It is meant to. I don't know how you've been learning and I have no proof that doing it differently will help, but I'm optimistic that it can. The main takeaway is this: be patient with yourself. Solving problems and building things is hard. It's ok to progress slowly. Don't try to skip ahead, especially early on.
(also this comment isn't directed at you specifically, but at anyone who shares your frustration)

I was gonna write an entire rant opposing the meme, but thought better of it as it seems most people here agree with me.
BUT I think that once you've got some basics down, there really is no better way to improve than to do. The key is to start at the appropriate level of complexity for your level of experience.
Obviously I don't know what that is for you specifically, but I think in general it's a good idea to start simple. Don't try to engineer an entire application as your first programming activity.

Find an easy (and simple! as in - a single function with well defined inputs and outputs and no side effects) problem; either think of something yourself, or pick an easy problem from an online platform like leetcode or codechef. And try to solve the problem yourself. There's no need to get stuck for ages, but give it an honest try.
I think a decent heuristic for determining if you have a useful problem is whether you feel like you've made significant progress towards a solution after an hour or two. If not, readjust and pick a different problem. There's no point in spending days on a problem that's not clicking for you.

If you weren't able to solve the problem, look at solutions. Pick one that seems most straight forward to you and try to understand it. When you think you do, give the original problem a little twist and try to solve that. While referencing the solution to the original if you need to.
If you're struggling with this kind of constrained problem, keep doing them. Seriously. Perhaps dial down the difficulty of the problems themselves until you can follow and understand the solutions. But keep struggling with trying to solve little problems from scratch. Because that's the essence of programming: you want the computer to do something and you need to figure out how to achieve that.
It's not automatic, intuitive, inspired creation. It's not magic. It's a difficult and uncertain process of exploration. I'm fairly confident that for most people, coding just isn't how their brain works, initially. And I'm also sure that for some it "clicks" much easier than for others. But fundamentally, the skill to code is like a muscle: it must be trained to be useful. You can listen to a hundred talks on the mechanics of bike riding, and be an expert on the physics. If you don't put in the hours on the pedals, you'll never be biking from A to B.
I think this period at the beginning is the most challenging and frustrating, because you're working so hard and seemingly progress so slowly. But the two are connected. You're not breezing through because it is hard. You're learning a new way of thinking. Everything else builds on this.

Once you're more comfortable with solving isolated problems like that, consider making a simple application. For example: read an input text file, replace all occurrences of one string with another string, write the resulting text to a new text file. Don't focus on perfection or best practices at first. Simply solve the problem the way you know how. Perhaps start with hard-coded values for the replacement, then make them configurable (e.g. by passing them as arguments to your application).

When you have a few small applications under your belt you can start to dream big. As in, start solving "real" problems. Like some automation that would help you or someone you know. Or tasks at work for a software company. Or that cool app you've always wanted to build. Working on real applications will give you more confidence and open the door to more learning. You'll run into lots of problems and learn how not to do things. So many ways not to do things.

TLDR: If it's not clicking, you need to, as a general rule, do less learning (in the conventional sense of absorbing and integrating information) and more doing. A lot of doing.