When it becomes 10x easier to express complex ideas and algorithms with a fraction of the code by making use of built-in language features, it's a breath of fresh air. The huge ecosystem of easy-to-use libraries for most anything doesn't hurt either. I can enjoy focusing on the problem I'm trying to solve instead of how to structure my code, what design pattern I should be using due to a limited static type system, how best to avoid memory corruption or seg faults, etc. It's akin to writing pseudo code, and then having it execute.
Another great thing about Python in particular is that it ends to be extremely easy to read other people's code due to semantic whitespace and also by language design of there only being one way to do things.
If Python turns out to be your thing, it's popular in data science/machine learning. If you lean in that direction, you may be able to work more on some complex math/stats problems and less on writing/maintaining large low-level codebases.
Also, that feeling that your code is not perfect is really common. I've been programming for ~20 years and it has never gone away. What's helped me a lot is focusing on the end result of the product, and its impact on the business and the value it creates. As someone who wants every line of code to be error free, as performant as possible, and properly structured, I find comfort knowing my "good enough" solution (and sometimes not-so-good solution) is making business owners and end-users more than happy.
Are you by any chance a perfectionist? There always will be places to improve your designs, especially if you're a perfectionist. Part of the answer is to not be such a perfectionist. You can scale perfection (or almost perfection) up to a point, but not beyond; 2000 lines seems like reasonably close to the breaking point.
Everybody's more comfortable in their preferred environment. Learning a new set of tools is not fun, because you're less effective than you would be in your normal environment, and you know it, and you compare your clumsiness with the new tool to how well you can do with the old one. But you sound like a pretty smart person, and you should be able to learn new things pretty readily. It sounds like you're unwilling to ride through the phase of uneasiness, of I-don't-know-how-to-do-anything-with-this-stupid-new-tool, of feeling dumb and clumsy, for long enough to become proficient. Look, I don't like having to do that, either. I usually kind of resist doing so for a while. But don't let it stop you from learning new stuff; whatever you go into, you're going to have to do that.
As for feeling dumb when learning new things; you may have to accept that no amount of experience/data point will let you "prove to yourself" that you will be good at a new challenge.
My software engineer/wife/best friend pretty much rocks at every job she's worked at. I've lost track of how many times she's told me "I don't know if I can do this job." when she starts a new job. She's been going through this cycle of new job=>self doubt=>mastery multiple times for over 2 decades and she still feels that way. Apparently, no amount of data points will dispel that feeling for some people.
http://www.quora.com/Reviews-of-Haskell-programming-language
From Tikhon Jelvis' answer:
"I personally rank languages on three largely orthogonal axes: productivity, maintainability and performance. That is: how quickly, comfortably and efficiently I can write code; how easy it is to ensure the code is correct, refactor it and add features; and how fast the resulting code can run. For different projects, the different axes will have different weights.
If the weights are about equal, Haskell is the best choice. Moreover, I do not know any language that can beat Haskell along any two axes: sure, there are (in fact, many) languages that perform better, and there are languages that lead to more correct, more maintainable code, and there are even languages that are more productive. But never more than one at a time."
I totally agree with the above and it's why I think Haskell is the best general programming language out there. I see it as the same niche as python, ruby, lisp, C#, Java, Go, etc.
If you think it a worthy goal to become good at it, best way is to do it more. Otherwise, stay to something you like and do well.