The question behind the question is whether statements from religious tradition should be taken seriously by scientists. These are statements made at a time when there were no human beings were aware of concepts such as galaxies. There are about 9,000 stars visible on a clear, moonless night, yet 2,000+ years ago the Talmud quotes a tradition that there are 10^18 stars.
Science is proud of its accuracy, yet when counting stars there is a great deal of guess work. Stars are counted in a small area and then extrapolated to be evenly distributed. Science estimates about 10^11 or 10^12. The key here is defining what is counted as a star. Next, galaxies are counted in a small area and extrapolated to be evenly distributed. Right now, that is also estimated to be about 10^11 to 10^12. The resulting estimate for the total number of stars is therefore 10^22 to 10^24.
(Just as an aside, if religious tradition made the claim that the average number of stars in a galaxy is almost exactly the number of galaxies in the universe I think science would have a problem with the coincidence.)
My feeling is that statements from religious tradition are like the answers in the back of a text book. If you work a problem and come up with a different answer then the first response should be to make sure that the problem was understood correctly, not that the answer given was a mistake. It is possible that the question is not the same. Here, perhaps the definition of a star is different or the definition of the size of the universe is different.
For the question of the size of the universe, the size of the "observable universe" changes as science progresses. Does that mean that the number from tradition is only valid for a specific level of scientific ability? Perhaps the number of stars will continue to increase and the traditional value will become progressively more wrong. That would not seem to be the universal result expected from religion. As a possible response, the "future visibility limit" may mean that the largest the number of observable stars can only ever be 2.36 times the current number of observable stars, which locks us in to about the current order of magnitude.
The number of grains of sand is often compared to the number of stars in the sky. I find one estimate of 10^18 and another of 10^24 (Sorry, but I am limited to 2 links, so you have to do the search yourself if you are interested). The 10^18 is based on grains of sand on the beach. This is not unreasonable, just a matter of definition. Perhaps the equivalent number for stars is based on a similar restrictive definition. In any case, religious traditional values do not seem to be out of line.
In the US (in 2009) there were 193 million "light duty vehicles, short wheel base," 40 million "light duty vehicles, long wheel base," 10 million trucks, and 9 million motorcycles. Using 10 million as the unit of measure, if asked the number of cars, either 19 (defined as cars) or 25 (defined as vehicles) would both be reasonable answers.
So, 2,000+ years ago nobody ever saw more than 10^4 stars, but the Talmud says that tradition puts the count as about 10^18. With all of our advances in science, without a clear definition of what is being counted, say that the number is more like 10^22 to 20^24.
This is supposed to discredit Rabbinic Torah?