You are right that somebody making $200/month in the US would be classified as extreme poor (and someone in India would not be considered extreme poor if he earns $200/month). I don't see the problem with this.
The world literacy rate and trend is supposedly pretty good. However, there are large differences in how literacy is measured and huge uncertainties in the numbers.
In some countries you're considered to be literate if you can sign your name, in others they just report 99-100% without doing any measurements.
UNESCO have plans for a better test to be applied globally, but I'm not sure any test that would lower the numbers would ever be welcomed.
It's not misleading. Malnutrition is going down, lack of education is going down, lack of access to healthcare is going down. The more you read the Guardian, the more you have misconceptions about how bad the world is developing.
That is intentionally misleading.
How about "extreme poverty" = "still suffering malnutrition" + "still lacking even basic education" + "still lacking any access to healthcare"
The $1.90 tells nothing and it is also dependant on purchasing power. If somebody was making $200/month in the US that would also be extreme poverty.