As interesting as this discussion is on our perceptions of Amtrak ridership in Atlanta, I think hard data is probably a better indication of the priorities public leaders place on Amtrak connectivity in Atlanta.
In 2013 there were 99,005 on/offs at the Atlanta station. This placed the Atlanta station at approximately the 70th position as far as volume. I could not compute Atlanta's exact rank because Amtrak data only shows ridership for the top 5 cities in each state and in a few states 5th busiest cities were still ahead of Atlanta.
Here is the data:
http://www.amtrak.com/ccurl/730/658/FY13-Record-Ridership-ATK-13-122.pdf
Of course what place Atlanta falls on Amtrak ridership with one route and without any commuter Amtrak compared to cities in Amtrak commuter networks is an apple and orange comparison. But, from the ridership data Atlanta is not a significant Amtrak market.
Two quick comparisons for these Amtrak statistics. Ridership on Marta in 2013 was 76.7 million.
http://www.itsmarta.com/kpichart_Rideship.aspx
Meanwhile, Harsfield-Jackson had over 94 Million passengers in 2013 -
http://www.atlanta-airport.com/docs/Traffic/201312.pdf
with 2.76 million passengers on the NY to Atlanta route -
http://www.transtats.bts.gov/.
This data probably confirms that Amtrak connectivity will be a low priority, in terms of transportation infrastructure, for the foreseeable future in Atlanta. That is not to say that if the Atlanta station had better connectivity and more routes that Amtrak ridership would not increase, its just reality of the situation. In a region where a penny sales tax for transportation improvements failed we are not going to see politicians prioritize Amtrak and spend a lot of money to improve the situation. Amtrak's not going to do it on its own when they have much larger ridership markets that need resources.
The most we can expect to see in the next 10-15 years is a new station (probably near Atlantic Station) with a Greyhound terminal and a few Marta bus routes. Linking Amtrak into Marta's rail lines seems very unlikely.