Kant is a hard man to understand, in part because his books are almost incomprehensible. You can start with the Critique of Pure Reason, but then you have a series of obstacles in your way that are truly challenging. That the real key to Kant lay beyond those initial stumbling blocks in the section on the ‘dialectic’ as with his antinomies always suggested an easier way to approach his thinking. But that has never been realized in any consistent course on Kant, although Schopenhauer in effect does this in his beautifully streamlined approach.
Whatever the case, it is completely normal to fail to understand Kant by reading him.
Whatever the case, there is hope (if you have tried and failed to read his critiques): the best way, if it can be found, is to arrive at an understanding of Kant BEFORE reading his books in detail. How to do that?
I can’t say, but in my own case I discovered Kant from the so-called ‘eonic model’, which is tantamount to entering his thought via the Third Antinomy, which ‘forces’ the issue of transcendental idealism.
In Schopenhauer there is not long lead up: you pull yourself up by your bootstraps and just enter the core of TI by sensing a kind of Upanishadic insight into representation, the terms ‘atman’ and ‘brahman’ being the lost rubble of ancient insights here, no long intelligible.
I think that the current debate over the Big Bang, with its religious and scientific poles, has already introduced most to the real crux of the Kantian dialectic, with its antinomy over a beginning in time. You might graduate from that hopeless debate to a ‘sudden Kantian’ glimpse of the reality pointed to beyond that set of contradictions, and thence come to understand the basic issue, then, and perhaps only then, starting to read Kant.
But, nota bene, those who understand Kant best don’t get there by reading him.