Think of a project you can build with the technology you want to learn, and then learn the technology through actually building the project.
I generally feel more motivated with this approach because at the end of the effort, I'll actually have something to show for my time. If the project idea is good enough, you might even be able to turn it into a business, but don't let the lack of good ideas stop you, because your primary goal is still learning, and for that, even clones of existing products can work (and you might be able to find a good twist on it that can differentiate your own project as you build it).
And for me personally, no amount of reading and exercises can compare with real usage if I wanted to become truly proficient at a technology.
But don't make it more perfect - do the next project. Make your focus to make some money, get more customers, etc.
The beautiful code will follow from that. It doesn't matter how beautiful the code is if no one uses it. (Well, maybe if you need to pass an exam or open source the code or something like that, slightly different goals for the code to 'succeed'.)