The options you have available to you depend on the contents of your txt files, but one general option (i.e., will work regardless) is to use a cell array:
p=dir('*.txt');
N=length(p);
A = cell(1,N); % cell array A
for k=1:N
% [fid]=fopen(p(k).name,'r');
fid = fopen(p(k).name,'r');
% A(k)=cell2mat(textscan(fid, '%f %f', 'HeaderLines', 4));
A{k} = cell2mat(textscan(fid, '%f %f', 'HeaderLines', 4)); % use {} for indexing A
% fid=fclose(fid);
fclose(fid);
end
Now each element of A is a cell containing the contents of one txt file. Access those contents again using curly braces {}:
A{1} % contents of the first file
A{2} % contents of the second file
A{end} % contents of the last file
% etc.
