That's about what I thought. I'd guess that most checks in the game are Average, with the occasional "hard" task at Challenging.
Here's some feedback: If that's an "average" character that you posted, shouldn't there be a difficulty around 50%?
That's the thing - the character isn't "average". It's a convenient illustrative profile because the numbers are easy to work with (50% chance of 1+ successes, 10% chance of 2 on each die).
And... even with that, I've made a mistake on the Challenging difficulty, as I forgot to account for the chance of rolling 2 successes on one die and none on the other. This pushes the chance to 39%.
Expertise 2 and Focus 2 is a fairly skilled character (skills range from 0-3 for both Expertise and Focus for most characters, with some characters having up to 5 for both in a small number of Signature Skills), so it isn't 'average'. Eight is an average attribute (normal range is 6-12 for most characters), yes, but the skills represent someone with a decent amount of training.
The only reason I picked those values was because they're convenient to work with.
Without it, it looks like this game will "live" at just the one difficulty, 75% (Average). Occasionally, something will be hard, and so the 25% (Challenging) kicks in. But, mostly, tasks will be Average.
You don't think that needs a little work?
Not really, because it comes down to design intent - the intended function of the rules - which I'll get to in a moment.
You have to pay Threat for extra dice.
Well, you have to find a way to get extra dice. Threat isn't the only way, though it is the universally-available one.
But the core this is true - you cannot reliably function at higher difficulties (or obtain useful quantities of Momentum) without obtaining additional dice. That's a feature, an intentional part of the design.
This is because a significant part of the design is not purely "how likely are you to succeed at X?" It also includes "how much are you willing to pay to succeed at X?"
As you say, the system is built assuming that some fundamental way of buying extra d20s is available.
In a lot of games where characters have access to a limited resource, that represents some additional factor on top of the baseline odds - if you're playing Mutants & Masterminds, use of a Hero Point (essentially, roll 1d10+10 instead of 1d20) is a way to circumvent the normal odds of the game occasionally. The assumed baseline of the game is rolling 1d20+modifier against target number.
2d20 does not have the same baseline assumption. The game is built with the assumption that characters will, at least occasionally, buy extra dice somehow. Consequently, the 'difficulty' of the game, absent the use of Threat, is lower than it might be if Threat were not a fundamental assumption (the majority of enemies are relatively weak - stronger foes often come with an inbuilt pool of 'personal Threat' that they can only use on themselves.
As a result, any variation of 2d20 that removed the Threat pool must replace it with something - hence the suggestion I made in an earlier post giving players a limited-but-quick-replenishing pool of Effort that lets them buy dice instead. This, however, removes the self-balancing relationship between player expenditure and GM resource.