If the ideal UX is one that doesn't make us think, and implicitly doesn't make us do things, delivering the ideal UX probably means predicting the experience people want, doing what they want, and reusing thinking they've done in the past. Where predictions and reuse don't remove the need to think, we can reduce and reward remaining thinking. While there are many frameworks for improving UX, I find myself mostly trying to predict, reuse, reduce, and reward thinking.
Predicting thinking
Most products are sold finished because suppliers predict or, if they have enough evidence, know buyers don't want to assemble their own products. Search engines predict, based on the few characters you've typed, what you want to search for. Predicting what people want, predicting what people will think given a set of choices, seems fundamental to product development. Unless what we want is to think, we'll wish to receive our wants without thinking or doing, and the best products will predict as much of this thinking for us and deliver the predictions best.
Reusing thinking
Metaphors are a plain example of thinking reuse. A metaphor makes existing intuition portable to something new, and products or features can reuse thinking this way too. For example, we group related information together because related, non-information things are often grouped together; a tree's trunk is found near its roots and branches, and the tree is spaced from neighboring trees, and thus a tree is this group of its trunk, roots, and branches. Reusing thinking can come naturally, but sometimes there isn't an obvious analog for our new thing. Still, it can help to find even the roughest analog because being forced to think a little is better than being forced to think a lot.
Reducing thinking
Technology often removes thinking and doing. Elastic slip-on shoes removed the need for us to tie shoelaces. Most of us still wear laced shoes, but I'd guess that's because shoelaces are traditional and fashionable. Many products don't suffer or benefit from forces like fashion, so when making the equivalent of a shoe, its usually best to consider them without laces. Removing product features or steps in a product experience can be as simple as forgetting the other shoes we've seen and remembering the ideal shoe performs the tasks of a shoe best; the ideal shoe probably isn't an existing one with slightly better laces.
Rewarding thinking
Sometimes, thinking isn't subject to prediction, reuse, or even reduction. Often, certain kinds of thinking are part or the point of a product. Having otherwise removed or reduced thinking, customers can fully attend to required or desired thinking, and while that's a significant benefit on its own, we can pursue opportunities to reward customers further. The message you want your product to send isn't only "we know what's important to you" but also "we're grateful for you" which often means making sacrifices beyond the product's scope or specification. A customer may be 100% satisfied with your product, but if you can provide satisfaction external to your product too, whether tangible or emotional, the total satisfaction your product provides can exceed 100%.
Introducing new thinking
I like to imagine the ideal UX creates a thinking surplus to be spent on introducing new thinking. New products often require new thinking, and we need customers to reserve enough thinking to experience the unique value our product delivers. The above rules can be applied to any thinking we imagine is new, but the unpredictable, irremovable, irreducible thinking that's left over is the anti-experience that's created when new experiences come into being. As you learn to predict, reuse, reduce, and reward this new thinking, you can introduce more new experiences and thinking. Hopefully, before long, you've created tons of accessible value that didn't exist before.
Forgive how naive this all sounds, and forgive my tone if it sounds authoritative. I haven't formally or even informally studied user experience. I'm mostly boxing up, in this abstract half-baked form, how I think about UX when I think about it to the extent that I can think about it. I'm not sure any of this is even directionally onto something, but it's what my mind produces when I try to put my feelings about UX into words. If I were better at writing, each sentence would have in it some form of "maybe" or "probably" and end with a question mark. So, imagine those are there.