When a person tells you - numbers are increasing, and you respond - "Have you factored in the increased amount of testing in your calculations?" - you're asking a basic question. This is like asking an engineer - did you consider static and dynamic loading? Do you really think that never occurred to him?? There's absolutely an implied questioning of basic competence in that question.
I think you're missing the two comparisons in the paper.
One is a study, rear-looking, trying to analyze observed information. They say, we saw a spread that looked like this, so based on what we know, the underlying dynamics of the spread must have looked like that. The window of observation closed on January 23. There's no prediction, only a conclusion. In order to see the spread we observed in this data set, there must have been more people who we didn't see. They don't conclude that these are asymptomatic. The word isn't even in the paper. They say "These undocumented infections often experience mild, limited or no symptoms and hence go unrecognized." That means unrecognized by the CDC or other groups, not the patients themselves. And they note "awareness among healthcare providers, public health officials and the availability of viral identification assays suggest that capacity for identifying previously missed infections has increased." And "...these measures are expected to increase reporting rates, reduce the proportion of undocumented infections, and decrease the growth and spread of infection. Indeed, estimation of the epidemiological characteristics of the outbreak after 23 January in China, indicate that government control efforts and population awareness have reduced the rate of spread of the virus."
The other is a paper explicitly used to guide public policy, which takes several assumptions including a very broad range of exponential growth to try to evaluate options for containment. It is fundamentally forward looking. At this point you have to make some assumptions. One of the assumptions they made is that 2/3s of cases will be sufficiently symptomatic to self-isolate.
That's not contrary with the modeled rate of undetected cases in Wuhan in the initial outbreak. There are certainly people in the previous undetected number who, with the increased sensitivity to everything, will self-isolate out of an abundance of caution when they wouldn't have before.
One set is undetected by the CDC or other monitoring agency. The other is detected by the person themselves.