I have not read the new book yet, but I have read a lot of Singer's books and articles over the years. What frustrates me about the famine-relief argument he has been making since his famous 1971 article "Famine, Affluence, and Morality" is that, in all that time, he has not given serious consideration to any of the following: (1) what the causes of wealth are, (2) what the causes of poverty are, and (3) what the consequences--both economic and moral--would be if the governments of wealthier nations worldwide actually acted on the recommendations he makes for the redistribution of wealth.
A further frustration is that he does not take seriously the careful and sustained criticisms of his argument that have appeared over the years. Many authors--including, for full disclosure, myself*--have reviewed his arguments, the empirical evidence that bears on the issues, and the economic, political, and moral implications of his position. Singer only occasionally mentions these criticisms, and when he does he typically dismisses them in few sentences or a brief paragraph.
Helping the poor rise out of poverty is a central concern of political and economic thinkers across the political spectrum; the disputes are not over whether it is good to help, but rather over what the best means to help are. It is unfair to consider only those people who end up agreeing with Singer as working in good faith, and it is unproductive to condemn those who disagree as holding a 'shameful' position. Other authors, like Garrett Cullity, do a much better job taking opposing positions seriously and reveiwing them charitably.
*My paper "Limits on Our Obligation to Give" appeared in Public Affairs Quarterly 14, 3 (July 2000): 183-203; the paper is reprinted in Justice: An Anthology edited by Louis Pojman. I also examine Singer's arguments in chapters 4 and 5 of my Actual Ethics.